3.2. Optimization of the Chemical Modification Process of the Activated Silica Gel
Activated silica gel was modified chemically using 3-amino propyl triethoxy silane (APTS). During the modification, silane coupling agents that attain the amine functional groups are substituted by silanol groups present on the silica surface according to the mechanism of electrophilic proton substitution, as illustrated in Scheme 2
. In this mechanism, the contribution of the centers of this surface has been taken into account [10
Different factors will influence the chemical substitution process of the activated silica gel, such as the reaction time, the reactant concentration and the reaction temperature. These factors will be optimized to functionalize the activated silica gel with the amine functional group.
Firstly, the effect of reaction time on the sorption efficiency of the produced ANSG was monitored to determine the copper removal capacity of the modified silica gel produced at each studied reaction time interval (Figure 2
A). It was elucidated by this figure that as the reaction period increased, the copper removal capacity of the produced material was incremented. This may be related to the short time reaction intervals leading to an incomplete chemical modification process for the activated silica gel. Consequently, it was suggested that the optimum reaction time for the chemical medication of silica gel is 22 h. This reaction time interval is compatible with many researchers’ results [11
], where they stated that 24 h is a sufficient period for completing the chemical modification process of silica gel [9
The solvent has a critical role in the reaction rate because solvents may or may not surround a nucleophile, thus hindering or not hindering its approach to the carbon atom [12
]. Therefore, the ratio of toluene solvent to silica gel will be firstly optimized to adjust the reactant ratio of the reaction mixture to prepare the most efficient sorbent material for copper sorption at the predetermined 22-h reaction time. The effect of the changing toluene to silica gel concentration from 5–60 mL/g at a fixed 3-amino propyl triethoxy silane (APTS) concentration was monitored for the copper sorption affinity of the produced ANSG. Figure 2
B demonstrates that the increment in the toluene concentration improves the copper removal affinity of the produced ANSG. This may return to the competition between both the solvent and organic groups for the reaction with the surface silanol groups present in the pores by means of an interaction with APTS generated through hydrogen bonds during the functionalization process. As the solvent concentration increased, a preference of hydrogen bonding occurs between the central region of APTS and the silanol groups of silica gel, rather than the weak van der Waals forces on the periphery. As such, the molecules of the APTS are more thermodynamically stabilized by hydrogen bonds than by the van der Waals interactions within the toluene solvent [13
]. Finally, it can be concluded that the optimum toluene to silica gel ratio was recorded to be 20 mL/g.
Based on this optimum toluene to silica gel ratio, the most proper amount of APTS-functionalized agent utilized for grafting the activated silica gel to produce a highly efficient copper sorption adsorbent material will be determined. The studied reaction ratio of APTS to silica gel in the reaction mixture was varied from 0.5–6 mL/g using 20 mL/g toluene solvent. The influence of APTS concentration on the copper sorption performance of the produced sorbent materials was illustrated in Figure 2
C. It was evident from this figure that the increment in the amount of APTS in the reaction mixture was associated with improvement in the copper removal capacity of the produced modified silica gel. This observation may return to incrementing in the degree of silica gel functionalized with amine groups as the amount of APTS coupling agent increased [13
]. That is, it improves the adsorption capacity of modified silica gel for copper removal. Accordingly, the optimum APTS to silica gel ratio was determined to be equal to 2.5 mL/g.
The effect of the reaction temperature on the silica gel amine grafting process was elucidated using all of the predetermined optimum silica gel modification process conditions. The copper removal capacity of the produced modified silica gel at different reaction temperatures was investigated at Figure 2
D. It is clear that the reaction temperature has a positive effect on the copper sorption affinity for the produced modified silica gel. This may be in regards to the fact that the increase in the reaction temperature favors the diffusion rate of reactants. That is, it improves the reaction rate and subsequently increases the degree of silica gel grafting by amine groups, which enhance the adsorption behavior of the modified silica gel to increase its affinity for copper removal. Accordingly, the optimum reaction temperature was 70 °C for the chemical modification process, where it produces adsorbent material with a 98% copper removal affinity.
According to the optimized preparation conditions, the most proper amine-functionalized silica gel (ANSG) that attains the optimum copper decontamination of 98% was synthesized through the co-condensation of 1 g silica gel with 20 mL toluene and 2.5 mL silane for 22 h at 70 °C. This functionalized silica gel was immobilized with nano-magnetite to attain nano-magnetic amine-functionalized silica gel using a co-precipitation technique.
3.3. Characterization of the Magnetic Amine-Functionalized Nano-Silica Gel
The properties of MANSG were mostly compared to its parent amine-functionalized silica gel (ANSG) before immobilization to record the main difference between the two matrices.
The FTIR spectra of both amine-functionalized silica gel and its magnetic immobilized matrices are investigated in Figure 3
to determine their main functional groups. Comparing the two spectrums, they show relatively similar spectra and reveal the presence of an amine-functionalized band (NH2
) at around 1650 cm−1
. Additionally, both matrices demonstrated the stretching band at 3200 cm−1
for functionalized silica gel that shifted to 2943.17 cm−1
for the magnetic silica gel, which was assigned to a C-H weak band of the carbon chain of the pendant group attached to the inorganic silica matrix [14
]. The shoulder band at 960 cm−1
present in the FTIR spectra of these matrices is associated to SiOH vibrations and composed of Si-O-H stretching. A new observed band was recorded in the magnetic immobilized silica gel spectrum around 750 cm−1
, which is responsible for the vibration of the M-O-M of the magnetite particle immobilized onto the silica gel matrix [15
]. These FTIR results confirm the amine functionalization of silica gel, as well as the immobilization of magnetite onto the functionalized silica gel.
exemplifies the XRD patterns of the mesoporous materials of the amine-functionalized silica gel (ANSG) and its magnetic matrices (MANSG). The two patterns appear to consist of a broad peak in a low 2θ angle range, which suggests the presence of ordering mesoporous structure materials. Meanwhile, the ANSG pattern (Figure 4
A) displayed a well-resolved pattern at low 2θ values with a sharp diffraction peak at 8.2° and a less intensive peak at the angle of 23° associated with (370), (222) reflections, respectively. The absence of peaks at a higher angle confirms that the silica layer has an amorphous structure [14
]. On the other hand, noticeable characteristic peaks were observed at high 2θ values for the XRD pattern (Figure 4
A) of magnetic-functionalized silica gel (ANSG) that explored the crystallinity of this matrix compared to the functionalized silica gel. Figure 4
B shows the sum of reflection intensities (red highlighted) at 2θ of 35.5°, 43°, 57° and 63°, which correspond to the (311), (400), (511) and (440) planes of the cubic crystal of iron oxide Fe3
, which confirmed by standard data for magnetite (Ref: 04007-1060) [16
]. These peaks confirm the presence of crystalline magnetite nanoparticles. However, the intensity of the immobilized magnetite characteristic peaks is not high enough compared to the pure magnetite reference (Ref: 04-007-1060). This may be in regards to the incorporation of some of magnetite nanoparticles within the pores of the silica gel mesoporous structure [16
represents the SEM micrographs of the amine-functionalized silica gel (Figure 5
A) and its magnetic immobilized hybrid (Figure 5
B). It was indicated that the particle morphology of the two matrices has spherical shapes with 100-nm and 50-nm average diameters for ANSG and MANSG, respectively. However, the raw silica gel before the chemical modification process has an average diameter equal to 20 µm (as indicated from its suppliers). Accordingly, these results designated that both the chemical functionalization and magnetite immobilization processes reduce the silica gel particle diameter from the micro-scale into the nano-scale. This may be owed to the severe chemical conditions of these processes that deform the silica gel bulk particles into nanospherical shapes as the action of chemical treatment processes under high temperature [17
]. Moreover, the magnetic immobilized silica gel hybrid was subjected to further chemical treatment conditions of the magnetite immobilization process after the chemical modification process of silica gel functionalization that may explain the reduction of the nanosize of the magnetic matrices and its uniform shape compared to the amine-functionalized one. Moreover, the immobilization of magnetite nanoparticles onto the silica gel matrix is confirmed in Figure 6
B. It was indicated from this figure that there are black spots of magnetic nanoparticles (highlighted in red) present over the silica gel matrix. This figure gives predictions that the mechanism of magnetite immobilization onto the silica gel matrix is based on the uniform dispersion of small nanoparticles of magnetite onto the porous structure of the functionalized silica gel matrix. This suggested mechanism is in accordance with other mechanisms described for the immobilization of magnetite onto the zeolite matrix to fabricate the magnetite zeolite nanocomposite [18
]. This result confirms the formation of the magnetic amine-functionalized silica gel hybrid.
The particle size reduction and the regularity of the magnetic amine-functionalized silica gel are confirmed in Figure 6
; where TEM imaging showed the reduction in the particles of the magnetic silica gel hybrid and its homogeneity in Figure 6
B compared to the amine-functionalized one in Figure 6
A. Moreover, the immobilization of magnetite nanoparticles onto the silica gel matrix is confirmed in Figure 6
B, where the black spots of the magnetic nanoparticles (highlighted in red) were present over the silica gel matrix, confirming the formation of the magnetic amine-functionalized silica gel hybrid. Finally, the conversion of bulk silica gel to the nanosilica gel gives the prediction that the amine-functionalized and magnetic immobilized silica gel materials present the potential to be employed as adsorbents for metal ion uptake. Moreover, the coupling agents of amine group adsorbed on the surface of silica gel induce hydrophobicity at the material surface. Therefore, the silica gel matrices’ surface assumes an organophilic character that enhances their performance as adsorbent materials [10
In order to investigate the magnetic properties of the fabricated magnetic amine-functionalized silica gel hybrid (MANSG), the magnetization curve of the material at room temperature was elucidated at Figure 7
. It was indicated that the magnetization hysteresis loop appears with an S-like shape; thus it does not display magnetic remanence. Accordingly, the magnetic silica gel was considered to be a super-paramagnetic matrix. It is observed from Figure 7
that the magnetization is completely saturated at a value of 22.775 emu/g for the MANSG matrix. The comparatively low value of the saturation magnetization (MS) may be in regards to the good dispersion of the magnetite nanoparticles in the silica gel matrices, which lead to the easiest formation of the magnetic domain [18
]. It can be readily observed that the smaller particle sizes exhibit a smaller value of Ms, as expected, due to the surface disorder and modified cationic distribution [19
The thermal profile of the fabricated MANSG hybrid is investigated in Figure 8
. The resulting TGA material profile (Figure 8
A) showed two main degradation steps at mid-points of 26–128 °C and 302–633 °C. The first gradual loss step assigned with a low percent of material weight loss equivalent to 7% corresponds to the loss of humidity and water contamination in the magnetic amine-functionalized silica gel matrix. The second degradation step represents the de-hydroxylation of the organic amine group, which was grafted onto the silica surface [20
]. Generally, the total weight losses with respect to the surface water and amine group degradation of the magnetic hybrid matrix are less than 15%. These results indicate the high thermal stability of the fabricated MANSG hybrid material up to 800 °C.
The DSC pattern of the MANSG hybrid material (Figure 8
B) verifies the previously discussed TGA results, where it shows two main endothermic peaks. The first broad endothermic peak represented around 100 °C was attributed to the heat losses due to the gradual loss of external water molecules from the material. This is followed by a narrow shoulder, which terminates in another broad peak at a temperature of around 300 °C, which may be ascribed to the combustion of the amine group grafted onto the silica gel surface [20
3.4. Copper Sorption Profile of the Magnetic Amine-Functionalized Nano-Silica Gel
The copper sorption properties of the optimized amine-functionalized silica gel after nano-magnetite immobilization (MANSG) were investigated to explore the diversity of its behavior toward the copper ions.
Firstly, the copper sorption profiles of both magnetic amine-functionalized nano-silica gel hybrid (MANSG) and its parent activated silica gel before amine functionalization were compared to investigate the effect of amine group functionalization on the copper sorption process. On the other hand, a comparable investigation between the copper sorption behavior of both the magnetic hybrid (MANSG) and amine-functionalized silica gel before magnetite immobilization (ANSG) showed the effect of the magnetite immobilization process. Figure 9
A implies that the copper decontamination of activated silica gel improved from 10%–98% for the fabricated MANSG. This confirms that the amine functional group has a great influence on the copper removal process of the magnetic silica gel hybrid. Figure 9
B evidences that the magnetic hybrid (MANSG) attains 80% copper removal compared to 72% copper removal using ANSG. Accordingly, the presence of magnetite in the amine-functionalized silica gel slightly increases the capacity of the hybrid for the copper sorption process. Finally, it was indicated from the comparable investigation (Figure 9
) that both MANSG and ANSG have a high copper sorption efficiency and compared to their parent material of activated silica gel. Furthermore, the sorption process onto these materials occurred rapidly and reached equilibrium after 90 min compared to the activated silica gel. This is likely due to the higher accessibility of copper analyte through the amine group present in both the amine-functionalized silica gel and its magnetic hybrid matrices [21
]. The copper decontamination process onto both MANSG and ANSG showed a fast rate of sorption during the first hour of the sorbate-sorbent contact, and the rate of copper percent removal became almost insignificant due to the quick exhaustion of the sorption active sites. Moreover, the rate of the percent of copper ion removal was higher in the beginning due to the larger surface area of the adsorbent being available for the sorption of the ions [22
The effect of adsorbent dosage on the adsorption process was examined at the equilibrium time. Figure 10
indicates that the percentage of copper ion removal increased from 62%–99.9% as the MANSG dosage increased from 5–60 g/L. This may be in regards to the increment in the number of sorption sites (functional groups) at the adsorbent surface, which increase by increasing the dose of the adsorbent material toward the fixed copper ion concentration. These functionalized chemical groups of MANSG were important in the formation of van der Waals bonding with copper ions. Therefore, the additional functional groups inside the magnetic silica gel hybrid played an essential role in metal ions’ binding during the adsorption process, in addition to the adsorption properties of immobilized nano-magnetite onto the silica gel. The optimum dosage of MANSG may be considered as 20 g/L [22
The effect of initial metal ion concentration depends on the immediate relation between the concentration of the ion and the available binding sites on an adsorbent surface. It is clear from Figure 11
that the increase in ionic strength of copper solution from 10–1000 ppm using a fixed MANSG amount decreases the magnetic hybrid efficiency from 89% down to 80%. Moreover, a dramatic decline in the percentage of copper decontamination from 80% down to 17% was noticed as the ionic strength of the copper solution increased above 1000 ppm, up to 5000 ppm, using a fixed MANSG amount. This may be due to the saturation of adsorption sites onto the MANSG surface with copper ions; where at a low copper concentration, there will be unoccupied active sites on the magnetic hybrid surface. As the initial copper concentration increases, the active sites required for copper ions’ adsorption will be lacking [22
The acidity of waste solution has two effects on metal ion adsorption onto a specific adsorbent material. Firstly, concerning the acidic media, the protons’ presence in the acidic solutions can protonate the binding sites of the chelating molecules. However, the hydroxide presence in the basic solutions may be complex and precipitates many metal ions [9
]. In order to elucidate the influence of this important parameter in the sorption process, the pH of the copper solution was varied over the 2–12 range through the batch procedure. Figure 12
declares that the adsorption percentage of Cu(II) sorption onto MANSG increases with the increase in the pH values for the studied pH range. This may be related to the fact that at a low pH, the MANSG surface tends to be more positive, and electrostatic repulsion with copper cations takes place, inhibiting the sorption process, resulting in decreasing copper sorption onto MANSG. Therefore, the optimum pH values for the maximum Cu(II) were at pH ≥ 7. In order to avoid hydrolyzing, which may cause copper ion precipitation at higher pH values (above seven), pH 7.0 was chosen as the optimum pH for screening the influence of the remaining processing parameters on the sorption process. Moreover, this pH simulates the natural pH of drinking water [24
]. This result of the optimum pH value for copper removal onto MANSG was confirmed by many other researchers [9
The effect of agitation speed on the percent of copper removal using MANSG was monitored. The percentage of copper removal seemed to be affected by the agitation speed for values between 0 and 200 rpm, thus confirming that the influence of external diffusion on the copper sorption kinetics plays a significant role; where the increase in the agitation speed decreases the boundary layer resistances, which improve the copper mass transfer into the bulk solution, increasing the copper sorption process. However, as the mixing rate was improved above 200 rpm, the percentage of copper removal was still almost constant. Accordingly, a 200-rpm shaking rate is sufficient to assure that the entirety of the surface binding sites onto MANSG is readily available for copper ions’ uptake and is selected as the optimum mixing speed [24
The effect of solution temperature on the copper sorption process is illustrated in Figure 13
. This figure displays that the adsorption process of copper ions onto MANSG may be an exothermic process. This sorption behavior may tend to decrease the copper sorption onto MANSG as the solution temperature improved above 25 °C. This decline in the copper sorption with the increment in temperature may decrease the adsorptive forces between the metal ions and the active sites on the MANSG adsorbent surface as a result of decreasing its adsorption capacity [22